الفهرس 
CHAPTER 1Only 14 pages are availabe for public view
 |  | from | 240 |  |  |
| | | from | 240 | | |
Abstract
The first part of the thesis investigates the optimal design of water distribution networks without storage capacity using mathematical optimization. Both micro-genetic algorithm and hydraulic simulation of water network are used to find the optimal diameters of pipes of the network. The single-objective function of the optimization problem is the total cost of the capital costs of the piping network and the pump, and the operating cost of the pump, under the fulfillment of minimum nodal pressure requirements. An optimization methodology, Genetic Algorithm for Pumps (GAPUMPS), is used to minimize energy consumption. The described procedure is demonstrated in four case studies for water distribution networks with or without pumps. Three case studies are already established in the literature as traditional benchmark networks : the two-loop network, the three-loop network without pump, and the three-loop network with centrifugal pump. The results showed good agreement with the literature. The micro-genetic algorithm used in this study obtained better, or at least similar, optimal costs in the number of function evaluations compared with those of other evolutionary algorithms reported in the literature. The rest of the study deals with jet pump and jet-centrifugal pumpperformance.Jet pump plays a major role in high lift pumping, booster pumping, dredging, and transporting materials. The poor efficiency and the usually sharp peak in the efficiency–flow rate ratio (M) curve are the main undesirable features of jet pump. The liquid-jet mixing phenomena is well occur in a water jet pump. This complex mixing processes require accurate prediction methods of water jet pumps performance. The objective of the present mathematical simulation and optimization study is to investigate firstly the effects of parameters (e.g., the jet pump area ratio, nozzle position, and length of the mixing chamber) on the jet pump performance and secondly, to perform an optimization of water jet pumps taking jet pump efficiency as the objective function. A FORTRAN program was prepared for the hydraulic simulation and optimization of jet pumps called GAJETPUMP. In the optimization procedure, the genetic algorithm was used because it is a global optimization tool and provides multiple optimal solutions. The numerical optimization results obtained by the genetic algorithm were compared with those predicted by MATLAB and good agreement was observed.Moreover, the present study deals with water jet pump combined with centrifugal pump. The first part of this study focuses on the hydraulic simulation of a combined jet-centrifugal pump. A FORTRAN program was prepared for the hydraulic simulation called GAJETCENT. The second part focuses on the optimal design of a combined jet-centrifugal pump. For this purpose, the genetic algorithm optimization tool based on evolutionary algorithm is coupled with hydraulic simulation of water jet pump to evaluate its performance. The optimization cycle is performed by using both the hydraulic simulation of water jet pump and genetic algorithm in order to seek the maximum of the objective function which is increasing the efficiency of the combined jet-centrifugal pump by making some changes to the area ratio of the jet pump, R. The process ends with the reach of the predefined number of enerations.Organization of the thesis:This thesis is comprised of six chapters and five appendices as follows:Chapter Introductionhis chapter illustrates the outlines of some chosen topics related to water distribution network cost optimization, jet pump and jet-centrifugal pump performance.Chapter 2Literature ReviewThis chapter gives a thorough literature review on jet pump and centrifugal pump simulation and optimization with different research approaches.Chapter 3Mathematical ModelingThis chapter contains a full description and derivation of the mathematical modeling of water distribution network, and both the jet pump and combined jet-centrifugal pump.Chapter 4Genetic Algorithms and Optimization ProcedureThis chapter contains a detailed overview on the genetic algorithm and the optimization analysis of the water distribution network, jet pump and combined jet-centrifugal pump.Chapter 5Results and DiscussionThis chapter includes the detailed results of the simulation and optimization of water distribution network cost optimization, jet pump and jet-centrifugal pump performance.Chapter 6Conclusions and commendationsConclusions based on the present study are drawn and recommendations are mentioned.Appendix ACriteria of Acceptance or Rejection of M or R in MATLAB ResultsThis appendix provides a theoretical investigation of criteria of acceptance and rejection of the R or M values through the jet pump optimization analysis.Appendix BDerivation of Jet Pump Efficiency with Nozzle-to-Throat Spacing This appendix provides the detailed derivation of the jet pump with variable nozzle-to-throat spacing efficiency equation.Appendix CR-M Relationship when Jet Pump Efficiency Equal any ValueThis appendix provides the detailed erivation of the relation between R and M at any jet pump efficiency value.Appendix DR-M Relationship when Jet Pump Efficiency Equal ZeroThis appendix provides the detailed derivation of the relation between R and M when the jet pump efficiency equals zero.Appendix ESolution of Fourth Degree Polynomial Equation R-M RelationshipThis appendix provides the method used to solve the fourth degree polynomial R-M equation.